On Tue, 2007-06-05 at 13:24 -0500, William Marshall wrote:
> 
> We've had several servers end up with read-only JFS volumes shortly
> after reboot recently. I'm not sure how backlevel we are on jfs &
> jfsutils, but here's some data I gathered data on 2 of them. 
> 
> System 1: 
> Jun  1 20:11:21 rchs73gd kernel: ERROR: (device sd(8,17)): dbAllocAG:
> unable to allocate blocks 
> Jun  1 20:11:21 rchs73gd kernel: ERROR: (device sd(8,17)): remounting
> filesystem as read-only 

This happens when an allocation group (AG) claims to have free blocks,
but the lower-level code is unable to find any blocks to allocate.  It
would be difficult to figure out how the block map got messed up, as the
original corruption could have happened a while back.  Running fsck will
rebuild the block map and should fix the problem.  There's not much you
can do about it unless you see problems with this volume again.
> 
> Red Hat Enterprise Linux ES release 3 (Taroon Update 5) 
> jfsutils-1.1.7-2 
> Linux rchs73gd.xxx 2.4.21-47.0.1.ELsmp #1 SMP Fri Oct 13 17:56:20 EDT
> 2006 i686 i686 i386 GNU/Linux 
> This one is running a jfs.o that we built based on the "original" RH
> shipped JFS source + patches from about Jan 2006. 

I haven't looked at a 2.4 kernel in quite a while.  :-)  In addition,
Redhat was never very interested in keeping up-to-date with jfs patches.
> 
> System 2: (standard code from SuSE) 
> Jun  4 04:11:33 as1lnx2 kernel: ERROR: (device sdb1): XT_GETPAGE:
> xtree page corrupt 
> Jun  4 04:11:33 as1lnx2 kernel: ERROR: (device sdb1): XT_GETPAGE:
> xtree page corrupt 
> 
> # uname -a 
> Linux as1lnx2 2.6.5-7.282-pseries64 #1 SMP Tue Aug 29 10:40:40 UTC
> 2006 ppc64 ppc64 ppc64 GNU/Linux 
> # cat /etc/SuSE-release 
> SUSE LINUX Enterprise Server 9 (ppc) 
> VERSION = 9 
> PATCHLEVEL = 3 

This is still a bit old, so I'm not sure if this is caused by something
that has since been fixed.  Is your jfsutils older than 1.1.8?  That
fixed a problem replaying the journal that could cause this error.
> 
> On this one, fsck reported: 
> 
> **Phase 4 - Report Problems 
> File system object FF12387 is linked
> as: /lbmarsh/tsmacl_rst/tsmacl/aclfile 
> cannot repair the data format error(s) in this file. 
> cannot repair FF12387.  Will release. 

This is likely to be the same file that caused the warnings and
read-only mount.

Thanks,
Shaggy
-- 
David Kleikamp
IBM Linux Technology Center


-------------------------------------------------------------------------
This SF.net email is sponsored by DB2 Express
Download DB2 Express C - the FREE version of DB2 express and take
control of your XML. No limits. Just data. Click to get it now.
http://sourceforge.net/powerbar/db2/
_______________________________________________
Jfs-discussion mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/jfs-discussion

Reply via email to